States of Affairs, Maria Reicher, ed., (Frankfurt: Ontos Verlag 2009): 167-208. Defending Existentialism? Marian David University of Notre Dame This paper is concerned with a popular view about the nature of propositions, commonly known as the Russellian view of propositions. Alvin Plantinga has dubbed it, or more precisely, a crucial consequence of it, Existentialism, and in his paper “On Existentialism” (1983) he has presented a forceful argument intended as a reductio of this view. In what follows, I describe the main relevant ingredients of the Russellian view of propositions and states of affairs. I present a relatively simple response Russellians might want to make to Plantinga’s anti-existentialist argument. I then explore one aspect of this response—one that leads to some rather curious consequences for the Russellian view of propositions and states of affairs. 1. The Russellian View of Propositions and States of Affairs For present purposes, I take for granted that there are propositions. They are most perspicuously referred to by unadorned that-clauses, such as ‘that ants are insects’, or by what grammarians call appositive noun-phrases containing a that-clause, e.g. ‘the proposition that ants are insects’. Propositions are, as one says a bit metaphorically, the shareable contents of mental states and speech acts, such as believing, disbelieving, assuming, asserting, and denying. If I believe that ants are insects, then there is something I believe, namely (the proposition) that ants are insects. If you too believe that ants are insects, then there is something we both believe, namely (the proposition) that ants are insects. If our ancestors denied that ants are insects, then they denied what we believe, namely (the proposition) that ants are insects. Propositions are also bearers of truth and falsehood. Indeed, according to most of their advocates, propositions are the basic bearers of truth and falsehood: other states, acts, or things, such as beliefs, assumptions, assertions, utterances, and sentences, are called true or false, or correct or incorrect, only because they stand in certain relations to true or false propositions. My belief that ants are insects is true, or correct, because the proposition that is the content of this belief, the proposition that ants are insects, is true. If you assertively utter the sentence ‘Spiders are insects’, thereby asserting that spiders are insects, your assertion is false, or incorrect, or wrong, because the proposition that is the content of your assertion, the proposition that spiders are insects, is false; and the sentence you uttered is false too, because it expresses this same false proposition. Being bearers of truth or falsehood, propositions are also bearers of properties that spin-off from truth and falsehood. Most relevantly to our concerns will be their modal properties, such as being necessarily true, possibly true, contingently true, possibly false, etc. All of this is pretty much common ground among all advocates of propositions.1 Reflecting on the shareability of propositions between different persons at various places 1 But this use of the term ‘proposition’ is comparatively recent. Somewhat confusingly (to us), our ancestors used it in about the opposite way. As Norman Kretzmann (1970) informs us, that was because Boethius had defined the propositio as an oratio verum falsumve significans, that is, a speech signifying what is true or what is false (‘speech’ here includes spoken, written, and mental speech, i.e. thinking). Owing to Boethius’ authority, this became the standard usage of the term which can still be found in Mill’s Logic of 1872. Some time after that, e.g. in Moore’s 1899, ‘proposition’ switched sides, from the left of Boethius’ definition to the right, the term now being used to refer to the thing signified by what Boethius called a propositio. 1 States of Affairs, Maria Reicher, ed., (Frankfurt: Ontos Verlag 2009): 167-208. and times and on the objectivity of truth and falsehood, most advocates of propositions maintain that propositions are abstract beings. Propositions, they hold, are neither located in space nor undergoing any intrinsic qualitative changes through time; and they do not depend for their existence on actually being believed, disbelieved, assumed, asserted, or denied. Though we have discovered some truths, many still remain to be discovered; though we have succumbed to some falsehoods, many still remain to be succumbed to; and then there are many more truths and falsehoods, many more propositions, which in all likelihood no human being will ever believe, or disbelieve, or even so much as consider. Propositions, then, are often characterized as bearers of truth values that are possible contents of attitudes such as believing, disbelieving, assuming and considering. More precisely: x is a proposition iff x is necessarily such that x is true or false and it is possible that there be someone who believes or disbelieves or assumes or asserts or denies or entertains x. This characterization tells us a bit about what propositions are, or are supposed to be, but not really all that much. One might well think it mostly tells us about the role propositions are supposed to play in our overall view of mind, language, and the world: the role of truth-value bearers and possible contents of the so-called “propositional attitudes”. One might like to hear more: What is their inner nature? What, one might like to ask, are propositions made of? Not all advocates of propositions acknowledge this as a reasonable request. But some do. Russellians do. They tell us that propositions are structured complexes, things that consist of parts or constituents related to each other in certain describable ways. They will describe the many ways in which propositions may be composed of simpler constituents, including the ways in which the propositions familiar from Propositional Logic may be composed of simpler propositions together with certain relations (truth functions), as well as the ways in which the propositions familiar from Predicate Logic may be composed of properties and relations and particular things together with more relations (or functions).2 Russellians are very liberal when it comes to the question what sort of things propositions can be made of. There is only one general requirement on propositional constituents: every proposition must contain at least one n-place relation (a property counts as a 1-place relation), together with some other constituent or constituents that, as one says, plug into the slot or slots of the relation.3 There is no general restriction on what sorts of items can combine with an n-place relation to constitute a proposition. There are, for example, purely general propositions, such as the proposition that there are ants and the proposition that all ants are insects, which are constituted entirely by properties and relations (or functions). However, in the most basic sorts of propositions the items that combine with the n-place relation will be one or more particular things, such as bicycles, or Arnold Schwarzenegger, or ants. To put this a bit differently, according to the Russellians, there are not only linguistically singular sentences, i.e. sentences containing singular terms, there are also ontologically singular propositions, propositions containing particular things. So, when the eminent entomologist, Prof. Hölldobler, speaking of his current favorite, says, “Alma ist eine 2 The ‘is composed of’-locution is very much not to be taken to imply that propositions have to be put together by some process or some agent. Russellians might follow Russell (cf. 1918) and think of the well-formedness rules of “ideal languages”, such as the languages of Propositional and Predicate Logic, in two complementary ways. Looked at one way, the rules tell us how to put symbols together to generate well-formed formulas of the ideal language. Looked at the other way, they describe and exhibit the internal structure and composition of the propositions that are expressed by well-formed formulas. That’s why an ideal language is ideal. 3 The number of these other constituents, it seems, does not have to equal n, it can be smaller than n. Compare the proposition that Alma is biting Al with the proposition that Alma is biting Alma and also the proposition that Alma is biting herself. 2 States of Affairs, Maria Reicher, ed., (Frankfurt: Ontos Verlag 2009): 167-208. Blattschneiderameise”, the content of the thought he thereby expresses, i.e. the proposition that Alma is a leaf-cutter ant, has Alma, the ant herself, as a constituent. Frege, using a different example, took exception to this: “Mont Blanc with all its snowfields is not itself a component part of the thought that Mont Blanc is more than 4000 metres high” (Frege 1904: 163). To which Russell replied: “I believe that in spite of all its snowfields Mont Blanc itself is a component part of what is actually asserted in the proposition ‘Mont Blanc is more than 4000 metres high’” (Russell 1904: 169). Fregeans tend to find this absurd. Some reject, or at least are highly skeptical of, the very idea that propositions are structured complexes. They suspect that talk of propositions as having constituents and internal structure, taken from the world of concrete material things, is overly metaphorical and cannot be applied to abstract beings with any ontological seriousness. Others embrace the structural approach but reject the idea that just anything, including a contingent particular, can be a constituent of a proposition. They hold that all propositions are purely abstract beings, composed entirely of purely abstract beings, variously referred to as Fregean senses, or concepts (in the non-psychological sense), or pure properties and relations. All typical Fregeans, whether or not they are structuralists about propositions, maintain that all propositions are purely abstract and are, therefore, necessary beings which do not depend for their existence on any contingent beings.